No label colors not showing in Embeddings panel[BUG]

[pranavjadhav001]

Embeddings highlighted for each label/class using a different color not showing as expected, even after groundtruth.labels selection.

Code to reproduce issue

• Trained an embedding model on CUB-200-2011 Dataset
• Created a fiftyOne dataset using dataset type ImageClassificationDirectoryTree
• using Dataset filepaths, created embeddings array
• passed embeddings with dataset to compute_visualization fn, visualization method chosen 'umap' , dim 2
• launch_app either directly in jupyter cell or brower tab

System information

• OS Platform and Distribution : Linux Ubuntu 20.04
• Python version (python --version): 3.8.10
• FiftyOne version (fiftyone --version): 0.23.8
• FiftyOne installed from (pip or source): pip

[pranavjadhav001]

After looking more into it, the problem arises due to large no. of classes , for my experiment I'm using CUB-200-2011 Dataset which has 200 classes. When i reduced the classes to 50, the colors appeared. I've even tried to add custom color map with unique 200 colors, but that also doesn't seem to work.